Mirror mirror on the wall, what is the vainest metric of them all?
summary
Argues that headline numbers often mislead and gives concrete replacements: a PR blast that shows 100,000 page views → 20 downloads is vanity, so measure conversion rates; a build time drop from 1:26 to 0:45 is a tooling detail, not business impact; track cycle time reduced from ~10 weeks to minutes to find bottlenecks; collect operational metrics like mean time to resolution and crash rates, and measure repeatability and actionability before trusting any metric.
AI-written summary · about
chapters (7)
- 0:00Start
- 6:42Can I make a business decision based on this metric?
- 8:29Can I intentionally repeat this result?
- 9:09Is the data a real reflection of the truth?
- 11:01Number of deploys a day
- 34:13Cost of travel!
- 54:30Measure everything... right?
related talks
56m"Mirror mirror on the wall, what is the vainest metric of them all?
24mVanity performance metrics are worthless
06mDelivery Metrics: the good, the bad, and the totally ridiculous
26mMetrics and Modeling – Helping Teams See How to Improve
28mMetrics that Matter
01h 01mFinding Metrics that matter and using them safely
29mYour Metrics Are Lying: How to Stop Reporting and Start Learning
41mRethinking Measurement for High-Performing Teams